Related papers: Decoupling Generalizability and Membership Privacy Risks in Neural Networks

Decoupling Generalizability and Membership Privacy Risks in Neural Networks

URL: http://arxiv.org/abs/2602.02296v2
Date: Mon, 09 Feb 2026 18:44:22 GMT
Title: Decoupling Generalizability and Membership Privacy Risks in Neural Networks
Authors: Xingli Fang, Jung-Eun Kim,
Abstract summary: A deep learning model usually has to sacrifice some utilities when it acquires some other abilities or characteristics.<n>In this paper, we identify that the model's generalization and privacy risks exist in different regions in deep neural network architectures.<n>We propose Privacy-Preserving Training Principle (PPTP) to protect model components from privacy risks while minimizing the loss in generalizability.
Score: 8.210473195536077
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Abstract: A deep learning model usually has to sacrifice some utilities when it acquires some other abilities or characteristics. Privacy preservation has such trade-off relationships with utilities. The loss disparity between various defense approaches implies the potential to decouple generalizability and privacy risks to maximize privacy gain. In this paper, we identify that the model's generalization and privacy risks exist in different regions in deep neural network architectures. Based on the observations that we investigate, we propose Privacy-Preserving Training Principle (PPTP) to protect model components from privacy risks while minimizing the loss in generalizability. Through extensive evaluations, our approach shows significantly better maintenance in model generalizability while enhancing privacy preservation.

Related papers

Your Privacy Depends on Others: Collusion Vulnerabilities in Individual Differential Privacy [50.66105844449181]
Individual Differential Privacy (iDP) promises users control over their privacy, but this promise can be broken in practice.<n>We reveal a previously overlooked vulnerability in sampling-based iDP mechanisms.<n>We propose $(varepsilon_i,_i,overline)$-iDP a privacy contract that uses $$-divergences to provide users with a hard upper bound on their excess vulnerability.
arXiv Detail & Related papers (2026-01-19T10:26:12Z)
On the MIA Vulnerability Gap Between Private GANs and Diffusion Models [51.53790101362898]
Generative Adversarial Networks (GANs) and diffusion models have emerged as leading approaches for high-quality image synthesis.<n>We present the first unified theoretical and empirical analysis of the privacy risks faced by differentially private generative models.
arXiv Detail & Related papers (2025-09-03T14:18:22Z)
Privacy in Fine-tuning Large Language Models: Attacks, Defenses, and Future Directions [11.338466798715906]
Fine-tuning Large Language Models (LLMs) can achieve state-of-the-art performance across various domains.<n>This paper provides a comprehensive survey of privacy challenges associated with fine-tuning LLMs.<n>We highlight vulnerabilities to various privacy attacks, including membership inference, data extraction, and backdoor attacks.
arXiv Detail & Related papers (2024-12-21T06:41:29Z)
Bayes-Nash Generative Privacy Against Membership Inference Attacks [24.330984323956173]
We propose a game-theoretic framework modeling privacy protection as a Bayesian game between defender and attacker.<n>To address strategic complexity, we represent the defender's mixed strategy as a neural network generator mapping private datasets to public representations.<n>Our approach significantly outperforms state-of-the-art methods by generating stronger attacks and achieving better privacy-utility tradeoffs.
arXiv Detail & Related papers (2024-10-09T20:29:04Z)
Consistent community detection in multi-layer networks with heterogeneous differential privacy [4.451479907610764]
We propose a personalized edge flipping mechanism that allows data publishers to protect edge information based on each node's privacy preference. It can achieve differential privacy while preserving the community structure under the multi-layer degree-corrected block model. We show that better privacy protection of edges can be obtained for a proportion of nodes while allowing other nodes to give up their privacy.
arXiv Detail & Related papers (2024-06-20T22:49:55Z)
Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models [112.48136829374741]
In this paper, we unveil a new vulnerability: the privacy backdoor attack. When a victim fine-tunes a backdoored model, their training data will be leaked at a significantly higher rate than if they had fine-tuned a typical model. Our findings highlight a critical privacy concern within the machine learning community and call for a reevaluation of safety protocols in the use of open-source pre-trained models.
arXiv Detail & Related papers (2024-04-01T16:50:54Z)
Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks [72.51255282371805]
We prove a privacy bound for the KL divergence between model distributions on worst-case neighboring datasets. We find that this KL privacy bound is largely determined by the expected squared gradient norm relative to model parameters during training.
arXiv Detail & Related papers (2023-10-31T16:13:22Z)
How Do Input Attributes Impact the Privacy Loss in Differential Privacy? [55.492422758737575]
We study the connection between the per-subject norm in DP neural networks and individual privacy loss. We introduce a novel metric termed the Privacy Loss-Input Susceptibility (PLIS) which allows one to apportion the subject's privacy loss to their input attributes.
arXiv Detail & Related papers (2022-11-18T11:39:03Z)
Federated Deep Learning with Bayesian Privacy [28.99404058773532]
Federated learning (FL) aims to protect data privacy by cooperatively learning a model without sharing private data among users. Homomorphic encryption (HE) based methods provide secure privacy protections but suffer from extremely high computational and communication overheads. Deep learning with Differential Privacy (DP) was implemented as a practical learning algorithm at a manageable cost in complexity.
arXiv Detail & Related papers (2021-09-27T12:48:40Z)
Robustness Threats of Differential Privacy [70.818129585404]
We experimentally demonstrate that networks, trained with differential privacy, in some settings might be even more vulnerable in comparison to non-private versions. We study how the main ingredients of differentially private neural networks training, such as gradient clipping and noise addition, affect the robustness of the model.
arXiv Detail & Related papers (2020-12-14T18:59:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.